homelab/active/device_driveripper/driveripper.md

Driveripper

• Driveripper
  • General Principles
  • Important Locations
  • Monitoring Scripts
  • Quick Ansible Commands
  • Disk Mounts
  • Disk Performance Testing
  • General VM Notes
    • Best Practices
  • OSBuild Composer
  • Retired Disks
    • Sending emails

General Principles

1. Ingress: DDNS, HTTP, TLS, TCP/UDP Proxy
2. Backup: Borg, Storage, etc
3. Monitor: Glances, Emails, Scrubs, Defragmentation, PMLogger, etc.
4. Access: SSH, Cockpit
5. Virtualize: Libvirtd

Important Locations

• /etc/luks-keys: luks keys
• /usr/local/scripts: admin scripts

Monitoring Scripts

scp active/device_driveripper/btrfs-monitor.service driveripper:/etc/systemd/system/
scp active/device_driveripper/btrfs-monitor.timer driveripper:/etc/systemd/system/

systemctl enable --now btrfs-monitor.timer

Quick Ansible Commands

# Update Caddy
ansible-playbook -i ansible/inventory.yaml -l proxy active/podman_caddy/install_caddy.yaml

# Update DDNS
ansible-playbook -i ansible/inventory.yaml -l proxy active/podman_ddns/install_ddns.yaml

Disk Mounts

1. All btrfs subvolid=5 volumes should be mounted under /btrfs
2. Since /media is for removable media, and /mnt is for temporary mounts, all btrfs subvolumes should be mounted in /srv for consistency.

Disk Performance Testing

# Write
dd if=/dev/zero of=test.dat bs=1M count=10000 conv=fsync status=progress

# Read
dd if=test.dat of=/dev/null bs=1M count=10000 conv=fsync status=progress

General VM Notes

1. Use ext4 as your filesystem within the VM to cut down on multi COW
2. Use raw images (convert from qcow2 where possible) to cut down on multi COW
3. Create a folder at /srv/vm/boot for boot disks
4. Create a folder at /srv/vm/data for data disks

Best Practices

1. Remove the default bridge: virsh net-undefined default

OSBuild Composer

1. Blueprints are stored in /srv/smb/ducoterra/images/blueprints
2. Builds are stored in /srv/smb/ducoterra/images/builds

# Push a blueprint
composer-cli blueprints push /srv/smb/ducoterra/images/blueprints/fedora43-base.toml

# Start a build
composer-cli compose start fedora43-base qcow2

# Watch the build
watch composer-cli compose list running

# Remove the old image
mv /srv/smb/ducoterra/images/builds/fedora-43-base.qcow2 /srv/smb/ducoterra/images/builds/fedora-43-base-old.qcow2

# Extract the build
composer-cli compose image \
--filename /srv/smb/ducoterra/images/builds/fedora-43-base.qcow2 \
92a2f1dd-9238-4d50-ab1d-cd7c8e355cac

# Convert the build to raw
qemu-img convert -f qcow2 -O raw \
/srv/smb/pool0/ducoterra/images/builds/fedora-43-base.qcow2 \
/srv/vm/pool1/fedora-43-base.raw

# Install (Change password for default user ducoterra!)
virt-install \
--name "gitlab" \
--boot uefi,firmware.feature0.name=secure-boot,firmware.feature0.enabled=no \
--cpu host-passthrough --vcpus sockets=1,cores=8,threads=2 \
--ram=8192 \
--os-variant=fedora41 \
--network bridge:bridge0 \
--graphics none \
--console pty,target.type=virtio \
--import --disk "path=/srv/vm/pool1/gitlab.raw,bus=virtio"

# convert a cloud-init image to raw
qemu-img convert -f qcow2 -O raw \
/srv/smb/ducoterra/images/cloud/Fedora-Cloud-Base-Generic-43-1.6.x86_64.qcow2 \
/srv/vm/pool1/fedora-43-cloud.raw

# Resize
qemu-img resize -f raw /srv/vm/pool1/fedora-43-cloud.raw 128G

# Install with cloud-init
virt-install \
--name "freeipa" \
--boot uefi,firmware.feature0.name=secure-boot,firmware.feature0.enabled=no \
--cpu host-passthrough --vcpus sockets=1,cores=8,threads=2 \
--ram=8192 \
--os-variant=fedora41 \
--network bridge:bridge0 \
--graphics none \
--import --disk "path=/srv/vm/pool1/freeipa-boot.raw,bus=virtio" \
--cloud-init disable=yes,user-data="/srv/smb/ducoterra/images/cloud-init/fedora/cloud-init/user-data,meta-data=/srv/smb/ducoterra/images/cloud-init/fedora/cloud-init/meta-data"

# Reattach to console if needed
virsh console fedora-43-base-test

passwd ducoterra
hostnamectl hostname <hostname>

# Attaching an extra drive
fdisk /dev/vdb
> g
> n
> enter enter enter
> w

mkfs.ext4 /dev/vdb1
lsblk -fs # grab UUID
vim /etc/fstab
systemctl daemon-reload
mount -a --mkdir

Retired Disks

Retired 12-19-2025

Retired 12-19-2025

Sending emails

# s-nail is mailx
dnf install -y msmtp
vim /etc/ssmtp/ssmtp.conf

Fill out the configuration like this:

# A system wide configuration file is optional.
# If it exists, it usually defines a default account.
# This allows msmtp to be used like /usr/sbin/sendmail.
account default

# The SMTP smarthost
host stmp-server.com

# Use TLS on port 465. On this port, TLS starts without STARTTLS.
port 465
tls on
tls_starttls off

# Construct envelope-from addresses of the form "user@oursite.example"
from driveripper@reeselink.com
# Do not allow programs to override this envelope-from address via -f
allow_from_override off
# Always set a From header that matches the envelope-from address
set_from_header on

# Syslog logging with facility LOG_MAIL instead of the default LOG_USER
syslog LOG_MAIL

# User info
auth on
user username
password password

Send an email like this:

# Basic message
printf "Subject: Test\n\nhello there username." | msmtp admin@ducoterra.net

# Or some btrfs stats
printf "Subject: Pool0 Stats\n\n$(btrfs device stats /btrfs/pool0)" | msmtp admin@ducoterra.net

# Or some more btrfs stats
printf "Subject: Pool0 Stats\n\n$(btrfs filesystem show)" | msmtp admin@ducoterra.net

# Or some smartmontools stats
printf "To: admin@ducoterra.net\n\nFrom: driveripper@reeselink.com\n\nSubject: Disk Stats\n\n$(smartctl -a /dev/sda | grep "SMART Attributes" -A 18)" | msmtp admin@ducoterra.net